danielhanchen commited on
Commit
5fa96b5
·
verified ·
1 Parent(s): 8122bba

Upload folder using huggingface_hub

Browse files
config.json CHANGED
@@ -10,7 +10,6 @@
10
  "conf_conv_kernel_size": 5,
11
  "conf_num_attention_heads": 8,
12
  "conf_num_hidden_layers": 12,
13
- "conf_positional_bias_size": 256,
14
  "conf_reduction_factor": 4,
15
  "conf_residual_weight": 0.5,
16
  "gradient_clipping": 10000000000.0,
@@ -22,7 +21,6 @@
22
  128,
23
  32
24
  ],
25
- "sscp_conv_eps": 0.001,
26
  "sscp_conv_group_norm_eps": 0.001,
27
  "sscp_conv_kernel_size": [
28
  [
@@ -134,7 +132,6 @@
134
  "altup_active_idx": 0,
135
  "altup_coef_clip": 120.0,
136
  "altup_correct_scale": true,
137
- "altup_lr_multiplier": 1.0,
138
  "altup_num_inputs": 4,
139
  "attention_bias": false,
140
  "attention_dropout": 0.0,
@@ -225,7 +222,6 @@
225
  "num_hidden_layers": 35,
226
  "num_key_value_heads": 2,
227
  "num_kv_shared_layers": 15,
228
- "query_pre_attn_scalar": 256,
229
  "rms_norm_eps": 1e-06,
230
  "rope_local_base_freq": 10000.0,
231
  "rope_scaling": null,
@@ -237,11 +233,11 @@
237
  "vocab_size_per_layer_input": 262144
238
  },
239
  "torch_dtype": "bfloat16",
240
- "transformers_version": "4.53.0",
241
  "unsloth_fixed": true,
242
  "vision_config": {
243
  "architecture": "mobilenetv5_300m_enc",
244
- "do_pooling": true,
245
  "hidden_size": 2048,
246
  "initializer_range": 0.02,
247
  "label_names": [
 
10
  "conf_conv_kernel_size": 5,
11
  "conf_num_attention_heads": 8,
12
  "conf_num_hidden_layers": 12,
 
13
  "conf_reduction_factor": 4,
14
  "conf_residual_weight": 0.5,
15
  "gradient_clipping": 10000000000.0,
 
21
  128,
22
  32
23
  ],
 
24
  "sscp_conv_group_norm_eps": 0.001,
25
  "sscp_conv_kernel_size": [
26
  [
 
132
  "altup_active_idx": 0,
133
  "altup_coef_clip": 120.0,
134
  "altup_correct_scale": true,
 
135
  "altup_num_inputs": 4,
136
  "attention_bias": false,
137
  "attention_dropout": 0.0,
 
222
  "num_hidden_layers": 35,
223
  "num_key_value_heads": 2,
224
  "num_kv_shared_layers": 15,
 
225
  "rms_norm_eps": 1e-06,
226
  "rope_local_base_freq": 10000.0,
227
  "rope_scaling": null,
 
233
  "vocab_size_per_layer_input": 262144
234
  },
235
  "torch_dtype": "bfloat16",
236
+ "transformers_version": "4.53.1",
237
  "unsloth_fixed": true,
238
  "vision_config": {
239
  "architecture": "mobilenetv5_300m_enc",
240
+ "do_pooling": false,
241
  "hidden_size": 2048,
242
  "initializer_range": 0.02,
243
  "label_names": [
generation_config.json CHANGED
@@ -6,5 +6,5 @@
6
  "pad_token_id": 0,
7
  "top_k": 64,
8
  "top_p": 0.95,
9
- "transformers_version": "4.53.0"
10
  }
 
6
  "pad_token_id": 0,
7
  "top_k": 64,
8
  "top_p": 0.95,
9
+ "transformers_version": "4.53.1"
10
  }
model-00001-of-00003.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:7d721b20f0851c7b8d332008073930531b3f6dd62a49b5d4de1bd24d2a708e9c
3
  size 3723417839
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:42e91852f0a13c628882925014089a705ca7b297cacc77ea2dacf51d9ef53494
3
  size 3723417839
preprocessor_config.json CHANGED
@@ -41,7 +41,7 @@
41
  "processor_class": "Gemma3nProcessor",
42
  "resample": 2,
43
  "rescale_factor": 0.00392156862745098,
44
- "return_attention_mask": false,
45
  "return_tensors": null,
46
  "sampling_rate": 16000,
47
  "size": {
 
41
  "processor_class": "Gemma3nProcessor",
42
  "resample": 2,
43
  "rescale_factor": 0.00392156862745098,
44
+ "return_attention_mask": true,
45
  "return_tensors": null,
46
  "sampling_rate": 16000,
47
  "size": {